Commit | Line | Data |
---|---|---|
ef79bbde P |
1 | /*************************************************************************** |
2 | * PCSX-Revolution - PlayStation Emulator for Nintendo Wii * | |
3 | * Copyright (C) 2009-2010 PCSX-Revolution Dev Team * | |
4 | * <http://code.google.com/p/pcsx-revolution/> * | |
5 | * * | |
6 | * This program is free software; you can redistribute it and/or modify * | |
7 | * it under the terms of the GNU General Public License as published by * | |
8 | * the Free Software Foundation; either version 2 of the License, or * | |
9 | * (at your option) any later version. * | |
10 | * * | |
11 | * This program is distributed in the hope that it will be useful, * | |
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of * | |
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * | |
14 | * GNU General Public License for more details. * | |
15 | * * | |
16 | * You should have received a copy of the GNU General Public License * | |
17 | * along with this program; if not, write to the * | |
18 | * Free Software Foundation, Inc., * | |
19 | * 51 Franklin Street, Fifth Floor, Boston, MA 02111-1307 USA. * | |
20 | ***************************************************************************/ | |
21 | ||
22 | /* | |
23 | * GTE functions. | |
24 | */ | |
25 | ||
26 | #include "gte.h" | |
27 | #include "psxmem.h" | |
28 | ||
f74fb727 | 29 | #define VX(n) (n < 3 ? regs->CP2D.p[n << 1].sw.l : regs->CP2D.p[9].sw.l) |
30 | #define VY(n) (n < 3 ? regs->CP2D.p[n << 1].sw.h : regs->CP2D.p[10].sw.l) | |
31 | #define VZ(n) (n < 3 ? regs->CP2D.p[(n << 1) + 1].sw.l : regs->CP2D.p[11].sw.l) | |
32 | #define MX11(n) (n < 3 ? regs->CP2C.p[(n << 3)].sw.l : 0) | |
33 | #define MX12(n) (n < 3 ? regs->CP2C.p[(n << 3)].sw.h : 0) | |
34 | #define MX13(n) (n < 3 ? regs->CP2C.p[(n << 3) + 1].sw.l : 0) | |
35 | #define MX21(n) (n < 3 ? regs->CP2C.p[(n << 3) + 1].sw.h : 0) | |
36 | #define MX22(n) (n < 3 ? regs->CP2C.p[(n << 3) + 2].sw.l : 0) | |
37 | #define MX23(n) (n < 3 ? regs->CP2C.p[(n << 3) + 2].sw.h : 0) | |
38 | #define MX31(n) (n < 3 ? regs->CP2C.p[(n << 3) + 3].sw.l : 0) | |
39 | #define MX32(n) (n < 3 ? regs->CP2C.p[(n << 3) + 3].sw.h : 0) | |
40 | #define MX33(n) (n < 3 ? regs->CP2C.p[(n << 3) + 4].sw.l : 0) | |
41 | #define CV1(n) (n < 3 ? (s32)regs->CP2C.r[(n << 3) + 5] : 0) | |
42 | #define CV2(n) (n < 3 ? (s32)regs->CP2C.r[(n << 3) + 6] : 0) | |
43 | #define CV3(n) (n < 3 ? (s32)regs->CP2C.r[(n << 3) + 7] : 0) | |
44 | ||
45 | #define fSX(n) ((regs->CP2D.p)[((n) + 12)].sw.l) | |
46 | #define fSY(n) ((regs->CP2D.p)[((n) + 12)].sw.h) | |
47 | #define fSZ(n) ((regs->CP2D.p)[((n) + 17)].w.l) /* (n == 0) => SZ1; */ | |
48 | ||
49 | #define gteVXY0 (regs->CP2D.r[0]) | |
50 | #define gteVX0 (regs->CP2D.p[0].sw.l) | |
51 | #define gteVY0 (regs->CP2D.p[0].sw.h) | |
52 | #define gteVZ0 (regs->CP2D.p[1].sw.l) | |
53 | #define gteVXY1 (regs->CP2D.r[2]) | |
54 | #define gteVX1 (regs->CP2D.p[2].sw.l) | |
55 | #define gteVY1 (regs->CP2D.p[2].sw.h) | |
56 | #define gteVZ1 (regs->CP2D.p[3].sw.l) | |
57 | #define gteVXY2 (regs->CP2D.r[4]) | |
58 | #define gteVX2 (regs->CP2D.p[4].sw.l) | |
59 | #define gteVY2 (regs->CP2D.p[4].sw.h) | |
60 | #define gteVZ2 (regs->CP2D.p[5].sw.l) | |
61 | #define gteRGB (regs->CP2D.r[6]) | |
62 | #define gteR (regs->CP2D.p[6].b.l) | |
63 | #define gteG (regs->CP2D.p[6].b.h) | |
64 | #define gteB (regs->CP2D.p[6].b.h2) | |
65 | #define gteCODE (regs->CP2D.p[6].b.h3) | |
66 | #define gteOTZ (regs->CP2D.p[7].w.l) | |
67 | #define gteIR0 (regs->CP2D.p[8].sw.l) | |
68 | #define gteIR1 (regs->CP2D.p[9].sw.l) | |
69 | #define gteIR2 (regs->CP2D.p[10].sw.l) | |
70 | #define gteIR3 (regs->CP2D.p[11].sw.l) | |
71 | #define gteSXY0 (regs->CP2D.r[12]) | |
72 | #define gteSX0 (regs->CP2D.p[12].sw.l) | |
73 | #define gteSY0 (regs->CP2D.p[12].sw.h) | |
74 | #define gteSXY1 (regs->CP2D.r[13]) | |
75 | #define gteSX1 (regs->CP2D.p[13].sw.l) | |
76 | #define gteSY1 (regs->CP2D.p[13].sw.h) | |
77 | #define gteSXY2 (regs->CP2D.r[14]) | |
78 | #define gteSX2 (regs->CP2D.p[14].sw.l) | |
79 | #define gteSY2 (regs->CP2D.p[14].sw.h) | |
80 | #define gteSXYP (regs->CP2D.r[15]) | |
81 | #define gteSXP (regs->CP2D.p[15].sw.l) | |
82 | #define gteSYP (regs->CP2D.p[15].sw.h) | |
83 | #define gteSZ0 (regs->CP2D.p[16].w.l) | |
84 | #define gteSZ1 (regs->CP2D.p[17].w.l) | |
85 | #define gteSZ2 (regs->CP2D.p[18].w.l) | |
86 | #define gteSZ3 (regs->CP2D.p[19].w.l) | |
87 | #define gteRGB0 (regs->CP2D.r[20]) | |
88 | #define gteR0 (regs->CP2D.p[20].b.l) | |
89 | #define gteG0 (regs->CP2D.p[20].b.h) | |
90 | #define gteB0 (regs->CP2D.p[20].b.h2) | |
91 | #define gteCODE0 (regs->CP2D.p[20].b.h3) | |
92 | #define gteRGB1 (regs->CP2D.r[21]) | |
93 | #define gteR1 (regs->CP2D.p[21].b.l) | |
94 | #define gteG1 (regs->CP2D.p[21].b.h) | |
95 | #define gteB1 (regs->CP2D.p[21].b.h2) | |
96 | #define gteCODE1 (regs->CP2D.p[21].b.h3) | |
97 | #define gteRGB2 (regs->CP2D.r[22]) | |
98 | #define gteR2 (regs->CP2D.p[22].b.l) | |
99 | #define gteG2 (regs->CP2D.p[22].b.h) | |
100 | #define gteB2 (regs->CP2D.p[22].b.h2) | |
101 | #define gteCODE2 (regs->CP2D.p[22].b.h3) | |
102 | #define gteRES1 (regs->CP2D.r[23]) | |
103 | #define gteMAC0 (((s32 *)regs->CP2D.r)[24]) | |
104 | #define gteMAC1 (((s32 *)regs->CP2D.r)[25]) | |
105 | #define gteMAC2 (((s32 *)regs->CP2D.r)[26]) | |
106 | #define gteMAC3 (((s32 *)regs->CP2D.r)[27]) | |
107 | #define gteIRGB (regs->CP2D.r[28]) | |
108 | #define gteORGB (regs->CP2D.r[29]) | |
109 | #define gteLZCS (regs->CP2D.r[30]) | |
110 | #define gteLZCR (regs->CP2D.r[31]) | |
111 | ||
112 | #define gteR11R12 (((s32 *)regs->CP2C.r)[0]) | |
113 | #define gteR22R23 (((s32 *)regs->CP2C.r)[2]) | |
114 | #define gteR11 (regs->CP2C.p[0].sw.l) | |
115 | #define gteR12 (regs->CP2C.p[0].sw.h) | |
116 | #define gteR13 (regs->CP2C.p[1].sw.l) | |
117 | #define gteR21 (regs->CP2C.p[1].sw.h) | |
118 | #define gteR22 (regs->CP2C.p[2].sw.l) | |
119 | #define gteR23 (regs->CP2C.p[2].sw.h) | |
120 | #define gteR31 (regs->CP2C.p[3].sw.l) | |
121 | #define gteR32 (regs->CP2C.p[3].sw.h) | |
122 | #define gteR33 (regs->CP2C.p[4].sw.l) | |
123 | #define gteTRX (((s32 *)regs->CP2C.r)[5]) | |
124 | #define gteTRY (((s32 *)regs->CP2C.r)[6]) | |
125 | #define gteTRZ (((s32 *)regs->CP2C.r)[7]) | |
126 | #define gteL11 (regs->CP2C.p[8].sw.l) | |
127 | #define gteL12 (regs->CP2C.p[8].sw.h) | |
128 | #define gteL13 (regs->CP2C.p[9].sw.l) | |
129 | #define gteL21 (regs->CP2C.p[9].sw.h) | |
130 | #define gteL22 (regs->CP2C.p[10].sw.l) | |
131 | #define gteL23 (regs->CP2C.p[10].sw.h) | |
132 | #define gteL31 (regs->CP2C.p[11].sw.l) | |
133 | #define gteL32 (regs->CP2C.p[11].sw.h) | |
134 | #define gteL33 (regs->CP2C.p[12].sw.l) | |
135 | #define gteRBK (((s32 *)regs->CP2C.r)[13]) | |
136 | #define gteGBK (((s32 *)regs->CP2C.r)[14]) | |
137 | #define gteBBK (((s32 *)regs->CP2C.r)[15]) | |
138 | #define gteLR1 (regs->CP2C.p[16].sw.l) | |
139 | #define gteLR2 (regs->CP2C.p[16].sw.h) | |
140 | #define gteLR3 (regs->CP2C.p[17].sw.l) | |
141 | #define gteLG1 (regs->CP2C.p[17].sw.h) | |
142 | #define gteLG2 (regs->CP2C.p[18].sw.l) | |
143 | #define gteLG3 (regs->CP2C.p[18].sw.h) | |
144 | #define gteLB1 (regs->CP2C.p[19].sw.l) | |
145 | #define gteLB2 (regs->CP2C.p[19].sw.h) | |
146 | #define gteLB3 (regs->CP2C.p[20].sw.l) | |
147 | #define gteRFC (((s32 *)regs->CP2C.r)[21]) | |
148 | #define gteGFC (((s32 *)regs->CP2C.r)[22]) | |
149 | #define gteBFC (((s32 *)regs->CP2C.r)[23]) | |
150 | #define gteOFX (((s32 *)regs->CP2C.r)[24]) | |
151 | #define gteOFY (((s32 *)regs->CP2C.r)[25]) | |
b7159e1a | 152 | // senquack - gteH register is u16, not s16, and used in GTE that way. |
153 | // HOWEVER when read back by CPU using CFC2, it will be incorrectly | |
154 | // sign-extended by bug in original hardware, according to Nocash docs | |
155 | // GTE section 'Screen Offset and Distance'. The emulator does this | |
156 | // sign extension when it is loaded to GTE by CTC2. | |
157 | //#define gteH (psxRegs.CP2C.p[26].sw.l) | |
158 | #define gteH (psxRegs.CP2C.p[26].w.l) | |
f74fb727 | 159 | #define gteDQA (regs->CP2C.p[27].sw.l) |
160 | #define gteDQB (((s32 *)regs->CP2C.r)[28]) | |
161 | #define gteZSF3 (regs->CP2C.p[29].sw.l) | |
162 | #define gteZSF4 (regs->CP2C.p[30].sw.l) | |
163 | #define gteFLAG (regs->CP2C.r[31]) | |
ef79bbde P |
164 | |
165 | #define GTE_OP(op) ((op >> 20) & 31) | |
166 | #define GTE_SF(op) ((op >> 19) & 1) | |
167 | #define GTE_MX(op) ((op >> 17) & 3) | |
168 | #define GTE_V(op) ((op >> 15) & 3) | |
169 | #define GTE_CV(op) ((op >> 13) & 3) | |
170 | #define GTE_CD(op) ((op >> 11) & 3) /* not used */ | |
171 | #define GTE_LM(op) ((op >> 10) & 1) | |
172 | #define GTE_CT(op) ((op >> 6) & 15) /* not used */ | |
173 | #define GTE_FUNCT(op) (op & 63) | |
174 | ||
175 | #define gteop (psxRegs.code & 0x1ffffff) | |
176 | ||
59774ed0 | 177 | #ifndef FLAGLESS |
178 | ||
669fed93 | 179 | static inline s64 BOUNDS_(psxCP2Regs *regs, s64 n_value, s64 n_max, int n_maxflag, s64 n_min, int n_minflag) { |
ef79bbde P |
180 | if (n_value > n_max) { |
181 | gteFLAG |= n_maxflag; | |
182 | } else if (n_value < n_min) { | |
183 | gteFLAG |= n_minflag; | |
184 | } | |
185 | return n_value; | |
186 | } | |
187 | ||
f74fb727 | 188 | static inline s32 LIM_(psxCP2Regs *regs, s32 value, s32 max, s32 min, u32 flag) { |
ef79bbde P |
189 | s32 ret = value; |
190 | if (value > max) { | |
191 | gteFLAG |= flag; | |
192 | ret = max; | |
193 | } else if (value < min) { | |
194 | gteFLAG |= flag; | |
195 | ret = min; | |
196 | } | |
197 | return ret; | |
198 | } | |
199 | ||
f74fb727 | 200 | static inline u32 limE_(psxCP2Regs *regs, u32 result) { |
59774ed0 | 201 | if (result > 0x1ffff) { |
202 | gteFLAG |= (1 << 31) | (1 << 17); | |
203 | return 0x1ffff; | |
204 | } | |
205 | return result; | |
206 | } | |
207 | ||
208 | #else | |
209 | ||
f74fb727 | 210 | #define BOUNDS_(regs, a, ...) (a) |
59774ed0 | 211 | |
f74fb727 | 212 | static inline s32 LIM_(psxCP2Regs *regs, s32 value, s32 max, s32 min, u32 flag_unused) { |
59774ed0 | 213 | s32 ret = value; |
214 | if (value > max) | |
215 | ret = max; | |
216 | else if (value < min) | |
217 | ret = min; | |
218 | return ret; | |
219 | } | |
220 | ||
f74fb727 | 221 | static inline u32 limE_(psxCP2Regs *regs, u32 result) { |
bedfea38 | 222 | if (result > 0x1ffff) |
223 | return 0x1ffff; | |
224 | return result; | |
225 | } | |
59774ed0 | 226 | |
227 | #endif | |
228 | ||
f74fb727 | 229 | #define BOUNDS(n_value,n_max,n_maxflag,n_min,n_minflag) \ |
230 | BOUNDS_(regs,n_value,n_max,n_maxflag,n_min,n_minflag) | |
231 | #define LIM(value,max,min,flag) \ | |
232 | LIM_(regs,value,max,min,flag) | |
233 | #define limE(result) \ | |
234 | limE_(regs,result) | |
235 | ||
ef79bbde P |
236 | #define A1(a) BOUNDS((a), 0x7fffffff, (1 << 30), -(s64)0x80000000, (1 << 31) | (1 << 27)) |
237 | #define A2(a) BOUNDS((a), 0x7fffffff, (1 << 29), -(s64)0x80000000, (1 << 31) | (1 << 26)) | |
238 | #define A3(a) BOUNDS((a), 0x7fffffff, (1 << 28), -(s64)0x80000000, (1 << 31) | (1 << 25)) | |
239 | #define limB1(a, l) LIM((a), 0x7fff, -0x8000 * !l, (1 << 31) | (1 << 24)) | |
240 | #define limB2(a, l) LIM((a), 0x7fff, -0x8000 * !l, (1 << 31) | (1 << 23)) | |
241 | #define limB3(a, l) LIM((a), 0x7fff, -0x8000 * !l, (1 << 22)) | |
242 | #define limC1(a) LIM((a), 0x00ff, 0x0000, (1 << 21)) | |
243 | #define limC2(a) LIM((a), 0x00ff, 0x0000, (1 << 20)) | |
244 | #define limC3(a) LIM((a), 0x00ff, 0x0000, (1 << 19)) | |
245 | #define limD(a) LIM((a), 0xffff, 0x0000, (1 << 31) | (1 << 18)) | |
246 | ||
ef79bbde P |
247 | #define F(a) BOUNDS((a), 0x7fffffff, (1 << 31) | (1 << 16), -(s64)0x80000000, (1 << 31) | (1 << 15)) |
248 | #define limG1(a) LIM((a), 0x3ff, -0x400, (1 << 31) | (1 << 14)) | |
249 | #define limG2(a) LIM((a), 0x3ff, -0x400, (1 << 31) | (1 << 13)) | |
7384197d | 250 | #define limH(a) LIM((a), 0x1000, 0x0000, (1 << 12)) |
ef79bbde | 251 | |
6c6c9590 | 252 | #ifndef __arm__ |
253 | #define A1U A1 | |
254 | #define A2U A2 | |
255 | #define A3U A3 | |
256 | #else | |
257 | /* these are unlikely to be hit and usually waste cycles, don't want them on ARM */ | |
258 | #define A1U(x) (x) | |
259 | #define A2U(x) (x) | |
260 | #define A3U(x) (x) | |
261 | #endif | |
262 | ||
b7159e1a | 263 | |
264 | //senquack - n param should be unsigned (will be 'gteH' reg which is u16) | |
265 | #ifdef GTE_USE_NATIVE_DIVIDE | |
266 | INLINE u32 DIVIDE(u16 n, u16 d) { | |
267 | if (n < d * 2) { | |
268 | return ((u32)n << 16) / d; | |
269 | } | |
270 | return 0xffffffff; | |
271 | } | |
272 | #else | |
ef79bbde | 273 | #include "gte_divider.h" |
b7159e1a | 274 | #endif // GTE_USE_NATIVE_DIVIDE |
ef79bbde | 275 | |
59774ed0 | 276 | #ifndef FLAGLESS |
277 | ||
32631e6a | 278 | const unsigned char gte_cycletab[64] = { |
81dbbf4c | 279 | /* 1 2 3 4 5 6 7 8 9 a b c d e f */ |
280 | 0, 15, 0, 0, 0, 0, 8, 0, 0, 0, 0, 0, 6, 0, 0, 0, | |
281 | 8, 8, 8, 19, 13, 0, 44, 0, 0, 0, 0, 17, 11, 0, 14, 0, | |
282 | 30, 0, 0, 0, 0, 0, 0, 0, 5, 8, 17, 0, 0, 5, 6, 0, | |
283 | 23, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 5, 5, 39, | |
284 | }; | |
285 | ||
286 | // warning: called by the dynarec | |
287 | int gteCheckStallRaw(u32 op_cycles, psxRegisters *regs) { | |
288 | u32 left = regs->gteBusyCycle - regs->cycle; | |
289 | int stall = 0; | |
290 | ||
291 | if (left <= 44) { | |
292 | //printf("c %2u stall %2u %u\n", op_cycles, left, regs->cycle); | |
293 | regs->cycle = regs->gteBusyCycle; | |
294 | stall = left; | |
295 | } | |
296 | regs->gteBusyCycle = regs->cycle + op_cycles; | |
297 | return stall; | |
298 | } | |
299 | ||
300 | void gteCheckStall(u32 op) { | |
301 | gteCheckStallRaw(gte_cycletab[op], &psxRegs); | |
302 | } | |
303 | ||
ef79bbde | 304 | static inline u32 MFC2(int reg) { |
eac38522 | 305 | psxCP2Regs *regs = &psxRegs.CP2; |
ef79bbde P |
306 | switch (reg) { |
307 | case 1: | |
308 | case 3: | |
309 | case 5: | |
310 | case 8: | |
311 | case 9: | |
312 | case 10: | |
313 | case 11: | |
314 | psxRegs.CP2D.r[reg] = (s32)psxRegs.CP2D.p[reg].sw.l; | |
315 | break; | |
316 | ||
317 | case 7: | |
318 | case 16: | |
319 | case 17: | |
320 | case 18: | |
321 | case 19: | |
322 | psxRegs.CP2D.r[reg] = (u32)psxRegs.CP2D.p[reg].w.l; | |
323 | break; | |
324 | ||
325 | case 15: | |
326 | psxRegs.CP2D.r[reg] = gteSXY2; | |
327 | break; | |
328 | ||
329 | case 28: | |
ef79bbde P |
330 | case 29: |
331 | psxRegs.CP2D.r[reg] = LIM(gteIR1 >> 7, 0x1f, 0, 0) | | |
332 | (LIM(gteIR2 >> 7, 0x1f, 0, 0) << 5) | | |
333 | (LIM(gteIR3 >> 7, 0x1f, 0, 0) << 10); | |
334 | break; | |
335 | } | |
336 | return psxRegs.CP2D.r[reg]; | |
337 | } | |
338 | ||
339 | static inline void MTC2(u32 value, int reg) { | |
eac38522 | 340 | psxCP2Regs *regs = &psxRegs.CP2; |
ef79bbde P |
341 | switch (reg) { |
342 | case 15: | |
343 | gteSXY0 = gteSXY1; | |
344 | gteSXY1 = gteSXY2; | |
345 | gteSXY2 = value; | |
346 | gteSXYP = value; | |
347 | break; | |
348 | ||
349 | case 28: | |
350 | gteIRGB = value; | |
3968e69e | 351 | // not gteIR1 etc. just to be consistent with dynarec |
352 | regs->CP2D.n.ir1 = (value & 0x1f) << 7; | |
353 | regs->CP2D.n.ir2 = (value & 0x3e0) << 2; | |
354 | regs->CP2D.n.ir3 = (value & 0x7c00) >> 3; | |
ef79bbde P |
355 | break; |
356 | ||
357 | case 30: | |
358 | { | |
359 | int a; | |
360 | gteLZCS = value; | |
361 | ||
362 | a = gteLZCS; | |
363 | if (a > 0) { | |
364 | int i; | |
365 | for (i = 31; (a & (1 << i)) == 0 && i >= 0; i--); | |
366 | gteLZCR = 31 - i; | |
367 | } else if (a < 0) { | |
368 | int i; | |
369 | a ^= 0xffffffff; | |
370 | for (i = 31; (a & (1 << i)) == 0 && i >= 0; i--); | |
371 | gteLZCR = 31 - i; | |
372 | } else { | |
373 | gteLZCR = 32; | |
374 | } | |
375 | } | |
376 | break; | |
377 | ||
ef79bbde P |
378 | case 31: |
379 | return; | |
380 | ||
381 | default: | |
382 | psxRegs.CP2D.r[reg] = value; | |
383 | } | |
384 | } | |
385 | ||
386 | static inline void CTC2(u32 value, int reg) { | |
387 | switch (reg) { | |
388 | case 4: | |
389 | case 12: | |
390 | case 20: | |
391 | case 26: | |
392 | case 27: | |
393 | case 29: | |
394 | case 30: | |
395 | value = (s32)(s16)value; | |
396 | break; | |
397 | ||
398 | case 31: | |
399 | value = value & 0x7ffff000; | |
400 | if (value & 0x7f87e000) value |= 0x80000000; | |
401 | break; | |
402 | } | |
403 | ||
404 | psxRegs.CP2C.r[reg] = value; | |
405 | } | |
406 | ||
407 | void gteMFC2() { | |
408 | if (!_Rt_) return; | |
409 | psxRegs.GPR.r[_Rt_] = MFC2(_Rd_); | |
410 | } | |
411 | ||
412 | void gteCFC2() { | |
413 | if (!_Rt_) return; | |
414 | psxRegs.GPR.r[_Rt_] = psxRegs.CP2C.r[_Rd_]; | |
415 | } | |
416 | ||
417 | void gteMTC2() { | |
418 | MTC2(psxRegs.GPR.r[_Rt_], _Rd_); | |
419 | } | |
420 | ||
421 | void gteCTC2() { | |
422 | CTC2(psxRegs.GPR.r[_Rt_], _Rd_); | |
423 | } | |
424 | ||
425 | #define _oB_ (psxRegs.GPR.r[_Rs_] + _Imm_) | |
426 | ||
427 | void gteLWC2() { | |
428 | MTC2(psxMemRead32(_oB_), _Rt_); | |
429 | } | |
430 | ||
431 | void gteSWC2() { | |
432 | psxMemWrite32(_oB_, MFC2(_Rt_)); | |
433 | } | |
434 | ||
32631e6a | 435 | void gteLWC2_stall() { |
436 | gteCheckStall(0); | |
437 | gteLWC2(); | |
438 | } | |
439 | ||
440 | void gteSWC2_stall() { | |
441 | gteCheckStall(0); | |
442 | gteSWC2(); | |
443 | } | |
444 | ||
59774ed0 | 445 | #endif // FLAGLESS |
446 | ||
3ebefe71 | 447 | #if 0 |
a2544c92 | 448 | #define DIVIDE DIVIDE_ |
449 | static u32 DIVIDE_(s16 n, u16 d) { | |
450 | if (n >= 0 && n < d * 2) { | |
451 | s32 n_ = n; | |
452 | return ((n_ << 16) + d / 2) / d; | |
453 | //return (u32)((float)(n_ << 16) / (float)d + (float)0.5); | |
454 | } | |
455 | return 0xffffffff; | |
456 | } | |
3ebefe71 | 457 | #endif |
a2544c92 | 458 | |
f74fb727 | 459 | void gteRTPS(psxCP2Regs *regs) { |
ef79bbde | 460 | int quotient; |
8cb04d22 | 461 | s64 tmp; |
ef79bbde P |
462 | |
463 | #ifdef GTE_LOG | |
464 | GTE_LOG("GTE RTPS\n"); | |
465 | #endif | |
466 | gteFLAG = 0; | |
467 | ||
468 | gteMAC1 = A1((((s64)gteTRX << 12) + (gteR11 * gteVX0) + (gteR12 * gteVY0) + (gteR13 * gteVZ0)) >> 12); | |
469 | gteMAC2 = A2((((s64)gteTRY << 12) + (gteR21 * gteVX0) + (gteR22 * gteVY0) + (gteR23 * gteVZ0)) >> 12); | |
470 | gteMAC3 = A3((((s64)gteTRZ << 12) + (gteR31 * gteVX0) + (gteR32 * gteVY0) + (gteR33 * gteVZ0)) >> 12); | |
471 | gteIR1 = limB1(gteMAC1, 0); | |
472 | gteIR2 = limB2(gteMAC2, 0); | |
473 | gteIR3 = limB3(gteMAC3, 0); | |
474 | gteSZ0 = gteSZ1; | |
475 | gteSZ1 = gteSZ2; | |
476 | gteSZ2 = gteSZ3; | |
477 | gteSZ3 = limD(gteMAC3); | |
478 | quotient = limE(DIVIDE(gteH, gteSZ3)); | |
479 | gteSXY0 = gteSXY1; | |
480 | gteSXY1 = gteSXY2; | |
481 | gteSX2 = limG1(F((s64)gteOFX + ((s64)gteIR1 * quotient)) >> 16); | |
482 | gteSY2 = limG2(F((s64)gteOFY + ((s64)gteIR2 * quotient)) >> 16); | |
483 | ||
8cb04d22 | 484 | tmp = (s64)gteDQB + ((s64)gteDQA * quotient); |
485 | gteMAC0 = F(tmp); | |
486 | gteIR0 = limH(tmp >> 12); | |
ef79bbde P |
487 | } |
488 | ||
f74fb727 | 489 | void gteRTPT(psxCP2Regs *regs) { |
ef79bbde P |
490 | int quotient; |
491 | int v; | |
492 | s32 vx, vy, vz; | |
8cb04d22 | 493 | s64 tmp; |
ef79bbde P |
494 | |
495 | #ifdef GTE_LOG | |
496 | GTE_LOG("GTE RTPT\n"); | |
497 | #endif | |
498 | gteFLAG = 0; | |
499 | ||
500 | gteSZ0 = gteSZ3; | |
501 | for (v = 0; v < 3; v++) { | |
502 | vx = VX(v); | |
503 | vy = VY(v); | |
504 | vz = VZ(v); | |
505 | gteMAC1 = A1((((s64)gteTRX << 12) + (gteR11 * vx) + (gteR12 * vy) + (gteR13 * vz)) >> 12); | |
506 | gteMAC2 = A2((((s64)gteTRY << 12) + (gteR21 * vx) + (gteR22 * vy) + (gteR23 * vz)) >> 12); | |
507 | gteMAC3 = A3((((s64)gteTRZ << 12) + (gteR31 * vx) + (gteR32 * vy) + (gteR33 * vz)) >> 12); | |
508 | gteIR1 = limB1(gteMAC1, 0); | |
509 | gteIR2 = limB2(gteMAC2, 0); | |
510 | gteIR3 = limB3(gteMAC3, 0); | |
511 | fSZ(v) = limD(gteMAC3); | |
512 | quotient = limE(DIVIDE(gteH, fSZ(v))); | |
513 | fSX(v) = limG1(F((s64)gteOFX + ((s64)gteIR1 * quotient)) >> 16); | |
514 | fSY(v) = limG2(F((s64)gteOFY + ((s64)gteIR2 * quotient)) >> 16); | |
515 | } | |
8cb04d22 | 516 | |
517 | tmp = (s64)gteDQB + ((s64)gteDQA * quotient); | |
518 | gteMAC0 = F(tmp); | |
519 | gteIR0 = limH(tmp >> 12); | |
ef79bbde P |
520 | } |
521 | ||
f74fb727 | 522 | void gteMVMVA(psxCP2Regs *regs) { |
ef79bbde P |
523 | int shift = 12 * GTE_SF(gteop); |
524 | int mx = GTE_MX(gteop); | |
525 | int v = GTE_V(gteop); | |
526 | int cv = GTE_CV(gteop); | |
527 | int lm = GTE_LM(gteop); | |
528 | s32 vx = VX(v); | |
529 | s32 vy = VY(v); | |
530 | s32 vz = VZ(v); | |
531 | ||
532 | #ifdef GTE_LOG | |
533 | GTE_LOG("GTE MVMVA\n"); | |
534 | #endif | |
535 | gteFLAG = 0; | |
536 | ||
537 | gteMAC1 = A1((((s64)CV1(cv) << 12) + (MX11(mx) * vx) + (MX12(mx) * vy) + (MX13(mx) * vz)) >> shift); | |
538 | gteMAC2 = A2((((s64)CV2(cv) << 12) + (MX21(mx) * vx) + (MX22(mx) * vy) + (MX23(mx) * vz)) >> shift); | |
539 | gteMAC3 = A3((((s64)CV3(cv) << 12) + (MX31(mx) * vx) + (MX32(mx) * vy) + (MX33(mx) * vz)) >> shift); | |
540 | ||
541 | gteIR1 = limB1(gteMAC1, lm); | |
542 | gteIR2 = limB2(gteMAC2, lm); | |
543 | gteIR3 = limB3(gteMAC3, lm); | |
544 | } | |
545 | ||
f74fb727 | 546 | void gteNCLIP(psxCP2Regs *regs) { |
ef79bbde P |
547 | #ifdef GTE_LOG |
548 | GTE_LOG("GTE NCLIP\n"); | |
549 | #endif | |
550 | gteFLAG = 0; | |
551 | ||
552 | gteMAC0 = F((s64)gteSX0 * (gteSY1 - gteSY2) + | |
553 | gteSX1 * (gteSY2 - gteSY0) + | |
554 | gteSX2 * (gteSY0 - gteSY1)); | |
555 | } | |
556 | ||
f74fb727 | 557 | void gteAVSZ3(psxCP2Regs *regs) { |
ef79bbde P |
558 | #ifdef GTE_LOG |
559 | GTE_LOG("GTE AVSZ3\n"); | |
560 | #endif | |
561 | gteFLAG = 0; | |
562 | ||
b3254af9 | 563 | gteMAC0 = F((s64)gteZSF3 * (gteSZ1 + gteSZ2 + gteSZ3)); |
ef79bbde P |
564 | gteOTZ = limD(gteMAC0 >> 12); |
565 | } | |
566 | ||
f74fb727 | 567 | void gteAVSZ4(psxCP2Regs *regs) { |
ef79bbde P |
568 | #ifdef GTE_LOG |
569 | GTE_LOG("GTE AVSZ4\n"); | |
570 | #endif | |
571 | gteFLAG = 0; | |
572 | ||
b3254af9 | 573 | gteMAC0 = F((s64)gteZSF4 * (gteSZ0 + gteSZ1 + gteSZ2 + gteSZ3)); |
ef79bbde P |
574 | gteOTZ = limD(gteMAC0 >> 12); |
575 | } | |
576 | ||
f74fb727 | 577 | void gteSQR(psxCP2Regs *regs) { |
ef79bbde P |
578 | int shift = 12 * GTE_SF(gteop); |
579 | int lm = GTE_LM(gteop); | |
580 | ||
581 | #ifdef GTE_LOG | |
582 | GTE_LOG("GTE SQR\n"); | |
583 | #endif | |
584 | gteFLAG = 0; | |
585 | ||
b3254af9 | 586 | gteMAC1 = (gteIR1 * gteIR1) >> shift; |
587 | gteMAC2 = (gteIR2 * gteIR2) >> shift; | |
588 | gteMAC3 = (gteIR3 * gteIR3) >> shift; | |
7384197d | 589 | gteIR1 = limB1(gteMAC1, lm); |
590 | gteIR2 = limB2(gteMAC2, lm); | |
591 | gteIR3 = limB3(gteMAC3, lm); | |
ef79bbde P |
592 | } |
593 | ||
f74fb727 | 594 | void gteNCCS(psxCP2Regs *regs) { |
ef79bbde P |
595 | #ifdef GTE_LOG |
596 | GTE_LOG("GTE NCCS\n"); | |
597 | #endif | |
598 | gteFLAG = 0; | |
599 | ||
b3254af9 | 600 | gteMAC1 = ((s64)(gteL11 * gteVX0) + (gteL12 * gteVY0) + (gteL13 * gteVZ0)) >> 12; |
601 | gteMAC2 = ((s64)(gteL21 * gteVX0) + (gteL22 * gteVY0) + (gteL23 * gteVZ0)) >> 12; | |
602 | gteMAC3 = ((s64)(gteL31 * gteVX0) + (gteL32 * gteVY0) + (gteL33 * gteVZ0)) >> 12; | |
ef79bbde P |
603 | gteIR1 = limB1(gteMAC1, 1); |
604 | gteIR2 = limB2(gteMAC2, 1); | |
605 | gteIR3 = limB3(gteMAC3, 1); | |
606 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
607 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
608 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
609 | gteIR1 = limB1(gteMAC1, 1); | |
610 | gteIR2 = limB2(gteMAC2, 1); | |
611 | gteIR3 = limB3(gteMAC3, 1); | |
b3254af9 | 612 | gteMAC1 = ((s32)gteR * gteIR1) >> 8; |
613 | gteMAC2 = ((s32)gteG * gteIR2) >> 8; | |
614 | gteMAC3 = ((s32)gteB * gteIR3) >> 8; | |
615 | gteIR1 = gteMAC1; | |
616 | gteIR2 = gteMAC2; | |
617 | gteIR3 = gteMAC3; | |
ef79bbde P |
618 | |
619 | gteRGB0 = gteRGB1; | |
620 | gteRGB1 = gteRGB2; | |
621 | gteCODE2 = gteCODE; | |
622 | gteR2 = limC1(gteMAC1 >> 4); | |
623 | gteG2 = limC2(gteMAC2 >> 4); | |
624 | gteB2 = limC3(gteMAC3 >> 4); | |
625 | } | |
626 | ||
f74fb727 | 627 | void gteNCCT(psxCP2Regs *regs) { |
ef79bbde P |
628 | int v; |
629 | s32 vx, vy, vz; | |
630 | ||
631 | #ifdef GTE_LOG | |
632 | GTE_LOG("GTE NCCT\n"); | |
633 | #endif | |
634 | gteFLAG = 0; | |
635 | ||
636 | for (v = 0; v < 3; v++) { | |
637 | vx = VX(v); | |
638 | vy = VY(v); | |
639 | vz = VZ(v); | |
b3254af9 | 640 | gteMAC1 = ((s64)(gteL11 * vx) + (gteL12 * vy) + (gteL13 * vz)) >> 12; |
641 | gteMAC2 = ((s64)(gteL21 * vx) + (gteL22 * vy) + (gteL23 * vz)) >> 12; | |
642 | gteMAC3 = ((s64)(gteL31 * vx) + (gteL32 * vy) + (gteL33 * vz)) >> 12; | |
ef79bbde P |
643 | gteIR1 = limB1(gteMAC1, 1); |
644 | gteIR2 = limB2(gteMAC2, 1); | |
645 | gteIR3 = limB3(gteMAC3, 1); | |
646 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
647 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
648 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
649 | gteIR1 = limB1(gteMAC1, 1); | |
650 | gteIR2 = limB2(gteMAC2, 1); | |
651 | gteIR3 = limB3(gteMAC3, 1); | |
b3254af9 | 652 | gteMAC1 = ((s32)gteR * gteIR1) >> 8; |
653 | gteMAC2 = ((s32)gteG * gteIR2) >> 8; | |
654 | gteMAC3 = ((s32)gteB * gteIR3) >> 8; | |
ef79bbde P |
655 | |
656 | gteRGB0 = gteRGB1; | |
657 | gteRGB1 = gteRGB2; | |
658 | gteCODE2 = gteCODE; | |
659 | gteR2 = limC1(gteMAC1 >> 4); | |
660 | gteG2 = limC2(gteMAC2 >> 4); | |
661 | gteB2 = limC3(gteMAC3 >> 4); | |
662 | } | |
b3254af9 | 663 | gteIR1 = gteMAC1; |
664 | gteIR2 = gteMAC2; | |
665 | gteIR3 = gteMAC3; | |
ef79bbde P |
666 | } |
667 | ||
f74fb727 | 668 | void gteNCDS(psxCP2Regs *regs) { |
ef79bbde P |
669 | #ifdef GTE_LOG |
670 | GTE_LOG("GTE NCDS\n"); | |
671 | #endif | |
672 | gteFLAG = 0; | |
673 | ||
b3254af9 | 674 | gteMAC1 = ((s64)(gteL11 * gteVX0) + (gteL12 * gteVY0) + (gteL13 * gteVZ0)) >> 12; |
675 | gteMAC2 = ((s64)(gteL21 * gteVX0) + (gteL22 * gteVY0) + (gteL23 * gteVZ0)) >> 12; | |
676 | gteMAC3 = ((s64)(gteL31 * gteVX0) + (gteL32 * gteVY0) + (gteL33 * gteVZ0)) >> 12; | |
ef79bbde P |
677 | gteIR1 = limB1(gteMAC1, 1); |
678 | gteIR2 = limB2(gteMAC2, 1); | |
679 | gteIR3 = limB3(gteMAC3, 1); | |
680 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
681 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
682 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
683 | gteIR1 = limB1(gteMAC1, 1); | |
684 | gteIR2 = limB2(gteMAC2, 1); | |
685 | gteIR3 = limB3(gteMAC3, 1); | |
6c6c9590 | 686 | gteMAC1 = (((gteR << 4) * gteIR1) + (gteIR0 * limB1(A1U((s64)gteRFC - ((gteR * gteIR1) >> 8)), 0))) >> 12; |
687 | gteMAC2 = (((gteG << 4) * gteIR2) + (gteIR0 * limB2(A2U((s64)gteGFC - ((gteG * gteIR2) >> 8)), 0))) >> 12; | |
688 | gteMAC3 = (((gteB << 4) * gteIR3) + (gteIR0 * limB3(A3U((s64)gteBFC - ((gteB * gteIR3) >> 8)), 0))) >> 12; | |
ef79bbde P |
689 | gteIR1 = limB1(gteMAC1, 1); |
690 | gteIR2 = limB2(gteMAC2, 1); | |
691 | gteIR3 = limB3(gteMAC3, 1); | |
692 | ||
693 | gteRGB0 = gteRGB1; | |
694 | gteRGB1 = gteRGB2; | |
695 | gteCODE2 = gteCODE; | |
696 | gteR2 = limC1(gteMAC1 >> 4); | |
697 | gteG2 = limC2(gteMAC2 >> 4); | |
698 | gteB2 = limC3(gteMAC3 >> 4); | |
699 | } | |
700 | ||
f74fb727 | 701 | void gteNCDT(psxCP2Regs *regs) { |
ef79bbde P |
702 | int v; |
703 | s32 vx, vy, vz; | |
704 | ||
705 | #ifdef GTE_LOG | |
706 | GTE_LOG("GTE NCDT\n"); | |
707 | #endif | |
708 | gteFLAG = 0; | |
709 | ||
710 | for (v = 0; v < 3; v++) { | |
711 | vx = VX(v); | |
712 | vy = VY(v); | |
713 | vz = VZ(v); | |
b3254af9 | 714 | gteMAC1 = ((s64)(gteL11 * vx) + (gteL12 * vy) + (gteL13 * vz)) >> 12; |
715 | gteMAC2 = ((s64)(gteL21 * vx) + (gteL22 * vy) + (gteL23 * vz)) >> 12; | |
716 | gteMAC3 = ((s64)(gteL31 * vx) + (gteL32 * vy) + (gteL33 * vz)) >> 12; | |
ef79bbde P |
717 | gteIR1 = limB1(gteMAC1, 1); |
718 | gteIR2 = limB2(gteMAC2, 1); | |
719 | gteIR3 = limB3(gteMAC3, 1); | |
720 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
721 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
722 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
723 | gteIR1 = limB1(gteMAC1, 1); | |
724 | gteIR2 = limB2(gteMAC2, 1); | |
725 | gteIR3 = limB3(gteMAC3, 1); | |
6c6c9590 | 726 | gteMAC1 = (((gteR << 4) * gteIR1) + (gteIR0 * limB1(A1U((s64)gteRFC - ((gteR * gteIR1) >> 8)), 0))) >> 12; |
727 | gteMAC2 = (((gteG << 4) * gteIR2) + (gteIR0 * limB2(A2U((s64)gteGFC - ((gteG * gteIR2) >> 8)), 0))) >> 12; | |
728 | gteMAC3 = (((gteB << 4) * gteIR3) + (gteIR0 * limB3(A3U((s64)gteBFC - ((gteB * gteIR3) >> 8)), 0))) >> 12; | |
ef79bbde P |
729 | |
730 | gteRGB0 = gteRGB1; | |
731 | gteRGB1 = gteRGB2; | |
732 | gteCODE2 = gteCODE; | |
733 | gteR2 = limC1(gteMAC1 >> 4); | |
734 | gteG2 = limC2(gteMAC2 >> 4); | |
735 | gteB2 = limC3(gteMAC3 >> 4); | |
736 | } | |
737 | gteIR1 = limB1(gteMAC1, 1); | |
738 | gteIR2 = limB2(gteMAC2, 1); | |
739 | gteIR3 = limB3(gteMAC3, 1); | |
740 | } | |
741 | ||
f74fb727 | 742 | void gteOP(psxCP2Regs *regs) { |
ef79bbde P |
743 | int shift = 12 * GTE_SF(gteop); |
744 | int lm = GTE_LM(gteop); | |
745 | ||
746 | #ifdef GTE_LOG | |
747 | GTE_LOG("GTE OP\n"); | |
748 | #endif | |
749 | gteFLAG = 0; | |
750 | ||
b3254af9 | 751 | gteMAC1 = ((gteR22 * gteIR3) - (gteR33 * gteIR2)) >> shift; |
752 | gteMAC2 = ((gteR33 * gteIR1) - (gteR11 * gteIR3)) >> shift; | |
753 | gteMAC3 = ((gteR11 * gteIR2) - (gteR22 * gteIR1)) >> shift; | |
ef79bbde P |
754 | gteIR1 = limB1(gteMAC1, lm); |
755 | gteIR2 = limB2(gteMAC2, lm); | |
756 | gteIR3 = limB3(gteMAC3, lm); | |
757 | } | |
758 | ||
f74fb727 | 759 | void gteDCPL(psxCP2Regs *regs) { |
ef79bbde P |
760 | int lm = GTE_LM(gteop); |
761 | ||
b3254af9 | 762 | s32 RIR1 = ((s32)gteR * gteIR1) >> 8; |
763 | s32 GIR2 = ((s32)gteG * gteIR2) >> 8; | |
764 | s32 BIR3 = ((s32)gteB * gteIR3) >> 8; | |
ef79bbde P |
765 | |
766 | #ifdef GTE_LOG | |
767 | GTE_LOG("GTE DCPL\n"); | |
768 | #endif | |
769 | gteFLAG = 0; | |
770 | ||
6c6c9590 | 771 | gteMAC1 = RIR1 + ((gteIR0 * limB1(A1U((s64)gteRFC - RIR1), 0)) >> 12); |
772 | gteMAC2 = GIR2 + ((gteIR0 * limB1(A2U((s64)gteGFC - GIR2), 0)) >> 12); | |
773 | gteMAC3 = BIR3 + ((gteIR0 * limB1(A3U((s64)gteBFC - BIR3), 0)) >> 12); | |
ef79bbde P |
774 | |
775 | gteIR1 = limB1(gteMAC1, lm); | |
776 | gteIR2 = limB2(gteMAC2, lm); | |
777 | gteIR3 = limB3(gteMAC3, lm); | |
778 | ||
779 | gteRGB0 = gteRGB1; | |
780 | gteRGB1 = gteRGB2; | |
781 | gteCODE2 = gteCODE; | |
782 | gteR2 = limC1(gteMAC1 >> 4); | |
783 | gteG2 = limC2(gteMAC2 >> 4); | |
784 | gteB2 = limC3(gteMAC3 >> 4); | |
785 | } | |
786 | ||
f74fb727 | 787 | void gteGPF(psxCP2Regs *regs) { |
ef79bbde P |
788 | int shift = 12 * GTE_SF(gteop); |
789 | ||
790 | #ifdef GTE_LOG | |
791 | GTE_LOG("GTE GPF\n"); | |
792 | #endif | |
793 | gteFLAG = 0; | |
794 | ||
b3254af9 | 795 | gteMAC1 = (gteIR0 * gteIR1) >> shift; |
796 | gteMAC2 = (gteIR0 * gteIR2) >> shift; | |
797 | gteMAC3 = (gteIR0 * gteIR3) >> shift; | |
ef79bbde P |
798 | gteIR1 = limB1(gteMAC1, 0); |
799 | gteIR2 = limB2(gteMAC2, 0); | |
800 | gteIR3 = limB3(gteMAC3, 0); | |
801 | ||
802 | gteRGB0 = gteRGB1; | |
803 | gteRGB1 = gteRGB2; | |
804 | gteCODE2 = gteCODE; | |
805 | gteR2 = limC1(gteMAC1 >> 4); | |
806 | gteG2 = limC2(gteMAC2 >> 4); | |
807 | gteB2 = limC3(gteMAC3 >> 4); | |
808 | } | |
809 | ||
f74fb727 | 810 | void gteGPL(psxCP2Regs *regs) { |
ef79bbde P |
811 | int shift = 12 * GTE_SF(gteop); |
812 | ||
813 | #ifdef GTE_LOG | |
814 | GTE_LOG("GTE GPL\n"); | |
815 | #endif | |
816 | gteFLAG = 0; | |
817 | ||
818 | gteMAC1 = A1((((s64)gteMAC1 << shift) + (gteIR0 * gteIR1)) >> shift); | |
819 | gteMAC2 = A2((((s64)gteMAC2 << shift) + (gteIR0 * gteIR2)) >> shift); | |
820 | gteMAC3 = A3((((s64)gteMAC3 << shift) + (gteIR0 * gteIR3)) >> shift); | |
821 | gteIR1 = limB1(gteMAC1, 0); | |
822 | gteIR2 = limB2(gteMAC2, 0); | |
823 | gteIR3 = limB3(gteMAC3, 0); | |
824 | ||
825 | gteRGB0 = gteRGB1; | |
826 | gteRGB1 = gteRGB2; | |
827 | gteCODE2 = gteCODE; | |
828 | gteR2 = limC1(gteMAC1 >> 4); | |
829 | gteG2 = limC2(gteMAC2 >> 4); | |
830 | gteB2 = limC3(gteMAC3 >> 4); | |
831 | } | |
832 | ||
f74fb727 | 833 | void gteDPCS(psxCP2Regs *regs) { |
ef79bbde P |
834 | int shift = 12 * GTE_SF(gteop); |
835 | ||
836 | #ifdef GTE_LOG | |
837 | GTE_LOG("GTE DPCS\n"); | |
838 | #endif | |
839 | gteFLAG = 0; | |
840 | ||
6c6c9590 | 841 | gteMAC1 = ((gteR << 16) + (gteIR0 * limB1(A1U(((s64)gteRFC - (gteR << 4)) << (12 - shift)), 0))) >> 12; |
842 | gteMAC2 = ((gteG << 16) + (gteIR0 * limB2(A2U(((s64)gteGFC - (gteG << 4)) << (12 - shift)), 0))) >> 12; | |
843 | gteMAC3 = ((gteB << 16) + (gteIR0 * limB3(A3U(((s64)gteBFC - (gteB << 4)) << (12 - shift)), 0))) >> 12; | |
ef79bbde P |
844 | |
845 | gteIR1 = limB1(gteMAC1, 0); | |
846 | gteIR2 = limB2(gteMAC2, 0); | |
847 | gteIR3 = limB3(gteMAC3, 0); | |
848 | gteRGB0 = gteRGB1; | |
849 | gteRGB1 = gteRGB2; | |
850 | gteCODE2 = gteCODE; | |
851 | gteR2 = limC1(gteMAC1 >> 4); | |
852 | gteG2 = limC2(gteMAC2 >> 4); | |
853 | gteB2 = limC3(gteMAC3 >> 4); | |
854 | } | |
855 | ||
f74fb727 | 856 | void gteDPCT(psxCP2Regs *regs) { |
ef79bbde P |
857 | int v; |
858 | ||
859 | #ifdef GTE_LOG | |
860 | GTE_LOG("GTE DPCT\n"); | |
861 | #endif | |
862 | gteFLAG = 0; | |
863 | ||
864 | for (v = 0; v < 3; v++) { | |
6c6c9590 | 865 | gteMAC1 = ((gteR0 << 16) + (gteIR0 * limB1(A1U((s64)gteRFC - (gteR0 << 4)), 0))) >> 12; |
866 | gteMAC2 = ((gteG0 << 16) + (gteIR0 * limB1(A2U((s64)gteGFC - (gteG0 << 4)), 0))) >> 12; | |
867 | gteMAC3 = ((gteB0 << 16) + (gteIR0 * limB1(A3U((s64)gteBFC - (gteB0 << 4)), 0))) >> 12; | |
ef79bbde P |
868 | |
869 | gteRGB0 = gteRGB1; | |
870 | gteRGB1 = gteRGB2; | |
871 | gteCODE2 = gteCODE; | |
872 | gteR2 = limC1(gteMAC1 >> 4); | |
873 | gteG2 = limC2(gteMAC2 >> 4); | |
874 | gteB2 = limC3(gteMAC3 >> 4); | |
875 | } | |
876 | gteIR1 = limB1(gteMAC1, 0); | |
877 | gteIR2 = limB2(gteMAC2, 0); | |
878 | gteIR3 = limB3(gteMAC3, 0); | |
879 | } | |
880 | ||
f74fb727 | 881 | void gteNCS(psxCP2Regs *regs) { |
ef79bbde P |
882 | #ifdef GTE_LOG |
883 | GTE_LOG("GTE NCS\n"); | |
884 | #endif | |
885 | gteFLAG = 0; | |
886 | ||
b3254af9 | 887 | gteMAC1 = ((s64)(gteL11 * gteVX0) + (gteL12 * gteVY0) + (gteL13 * gteVZ0)) >> 12; |
888 | gteMAC2 = ((s64)(gteL21 * gteVX0) + (gteL22 * gteVY0) + (gteL23 * gteVZ0)) >> 12; | |
889 | gteMAC3 = ((s64)(gteL31 * gteVX0) + (gteL32 * gteVY0) + (gteL33 * gteVZ0)) >> 12; | |
ef79bbde P |
890 | gteIR1 = limB1(gteMAC1, 1); |
891 | gteIR2 = limB2(gteMAC2, 1); | |
892 | gteIR3 = limB3(gteMAC3, 1); | |
893 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
894 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
895 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
896 | gteIR1 = limB1(gteMAC1, 1); | |
897 | gteIR2 = limB2(gteMAC2, 1); | |
898 | gteIR3 = limB3(gteMAC3, 1); | |
899 | ||
900 | gteRGB0 = gteRGB1; | |
901 | gteRGB1 = gteRGB2; | |
902 | gteCODE2 = gteCODE; | |
903 | gteR2 = limC1(gteMAC1 >> 4); | |
904 | gteG2 = limC2(gteMAC2 >> 4); | |
905 | gteB2 = limC3(gteMAC3 >> 4); | |
906 | } | |
907 | ||
f74fb727 | 908 | void gteNCT(psxCP2Regs *regs) { |
ef79bbde P |
909 | int v; |
910 | s32 vx, vy, vz; | |
911 | ||
912 | #ifdef GTE_LOG | |
913 | GTE_LOG("GTE NCT\n"); | |
914 | #endif | |
915 | gteFLAG = 0; | |
916 | ||
917 | for (v = 0; v < 3; v++) { | |
918 | vx = VX(v); | |
919 | vy = VY(v); | |
920 | vz = VZ(v); | |
b3254af9 | 921 | gteMAC1 = ((s64)(gteL11 * vx) + (gteL12 * vy) + (gteL13 * vz)) >> 12; |
922 | gteMAC2 = ((s64)(gteL21 * vx) + (gteL22 * vy) + (gteL23 * vz)) >> 12; | |
923 | gteMAC3 = ((s64)(gteL31 * vx) + (gteL32 * vy) + (gteL33 * vz)) >> 12; | |
ef79bbde P |
924 | gteIR1 = limB1(gteMAC1, 1); |
925 | gteIR2 = limB2(gteMAC2, 1); | |
926 | gteIR3 = limB3(gteMAC3, 1); | |
927 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
928 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
929 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
930 | gteRGB0 = gteRGB1; | |
931 | gteRGB1 = gteRGB2; | |
932 | gteCODE2 = gteCODE; | |
933 | gteR2 = limC1(gteMAC1 >> 4); | |
934 | gteG2 = limC2(gteMAC2 >> 4); | |
935 | gteB2 = limC3(gteMAC3 >> 4); | |
936 | } | |
937 | gteIR1 = limB1(gteMAC1, 1); | |
938 | gteIR2 = limB2(gteMAC2, 1); | |
939 | gteIR3 = limB3(gteMAC3, 1); | |
940 | } | |
941 | ||
f74fb727 | 942 | void gteCC(psxCP2Regs *regs) { |
ef79bbde P |
943 | #ifdef GTE_LOG |
944 | GTE_LOG("GTE CC\n"); | |
945 | #endif | |
946 | gteFLAG = 0; | |
947 | ||
948 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
949 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
950 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
951 | gteIR1 = limB1(gteMAC1, 1); | |
952 | gteIR2 = limB2(gteMAC2, 1); | |
953 | gteIR3 = limB3(gteMAC3, 1); | |
b3254af9 | 954 | gteMAC1 = ((s32)gteR * gteIR1) >> 8; |
955 | gteMAC2 = ((s32)gteG * gteIR2) >> 8; | |
956 | gteMAC3 = ((s32)gteB * gteIR3) >> 8; | |
ef79bbde P |
957 | gteIR1 = limB1(gteMAC1, 1); |
958 | gteIR2 = limB2(gteMAC2, 1); | |
959 | gteIR3 = limB3(gteMAC3, 1); | |
960 | ||
961 | gteRGB0 = gteRGB1; | |
962 | gteRGB1 = gteRGB2; | |
963 | gteCODE2 = gteCODE; | |
964 | gteR2 = limC1(gteMAC1 >> 4); | |
965 | gteG2 = limC2(gteMAC2 >> 4); | |
966 | gteB2 = limC3(gteMAC3 >> 4); | |
967 | } | |
968 | ||
f74fb727 | 969 | void gteINTPL(psxCP2Regs *regs) { |
ef79bbde P |
970 | int shift = 12 * GTE_SF(gteop); |
971 | int lm = GTE_LM(gteop); | |
972 | ||
973 | #ifdef GTE_LOG | |
974 | GTE_LOG("GTE INTPL\n"); | |
975 | #endif | |
976 | gteFLAG = 0; | |
977 | ||
6c6c9590 | 978 | gteMAC1 = ((gteIR1 << 12) + (gteIR0 * limB1(A1U((s64)gteRFC - gteIR1), 0))) >> shift; |
979 | gteMAC2 = ((gteIR2 << 12) + (gteIR0 * limB2(A2U((s64)gteGFC - gteIR2), 0))) >> shift; | |
980 | gteMAC3 = ((gteIR3 << 12) + (gteIR0 * limB3(A3U((s64)gteBFC - gteIR3), 0))) >> shift; | |
ef79bbde P |
981 | gteIR1 = limB1(gteMAC1, lm); |
982 | gteIR2 = limB2(gteMAC2, lm); | |
983 | gteIR3 = limB3(gteMAC3, lm); | |
984 | gteRGB0 = gteRGB1; | |
985 | gteRGB1 = gteRGB2; | |
986 | gteCODE2 = gteCODE; | |
987 | gteR2 = limC1(gteMAC1 >> 4); | |
988 | gteG2 = limC2(gteMAC2 >> 4); | |
989 | gteB2 = limC3(gteMAC3 >> 4); | |
990 | } | |
991 | ||
f74fb727 | 992 | void gteCDP(psxCP2Regs *regs) { |
ef79bbde P |
993 | #ifdef GTE_LOG |
994 | GTE_LOG("GTE CDP\n"); | |
995 | #endif | |
996 | gteFLAG = 0; | |
997 | ||
998 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); | |
999 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); | |
1000 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); | |
1001 | gteIR1 = limB1(gteMAC1, 1); | |
1002 | gteIR2 = limB2(gteMAC2, 1); | |
1003 | gteIR3 = limB3(gteMAC3, 1); | |
6c6c9590 | 1004 | gteMAC1 = (((gteR << 4) * gteIR1) + (gteIR0 * limB1(A1U((s64)gteRFC - ((gteR * gteIR1) >> 8)), 0))) >> 12; |
1005 | gteMAC2 = (((gteG << 4) * gteIR2) + (gteIR0 * limB2(A2U((s64)gteGFC - ((gteG * gteIR2) >> 8)), 0))) >> 12; | |
1006 | gteMAC3 = (((gteB << 4) * gteIR3) + (gteIR0 * limB3(A3U((s64)gteBFC - ((gteB * gteIR3) >> 8)), 0))) >> 12; | |
ef79bbde P |
1007 | gteIR1 = limB1(gteMAC1, 1); |
1008 | gteIR2 = limB2(gteMAC2, 1); | |
1009 | gteIR3 = limB3(gteMAC3, 1); | |
1010 | ||
1011 | gteRGB0 = gteRGB1; | |
1012 | gteRGB1 = gteRGB2; | |
1013 | gteCODE2 = gteCODE; | |
1014 | gteR2 = limC1(gteMAC1 >> 4); | |
1015 | gteG2 = limC2(gteMAC2 >> 4); | |
1016 | gteB2 = limC3(gteMAC3 >> 4); | |
1017 | } | |
6c0eefaf | 1018 | |
1019 | /* decomposed/parametrized versions for the recompiler */ | |
1020 | ||
1021 | #ifndef FLAGLESS | |
1022 | ||
1023 | void gteSQR_part_noshift(psxCP2Regs *regs) { | |
1024 | gteFLAG = 0; | |
1025 | ||
1026 | gteMAC1 = gteIR1 * gteIR1; | |
1027 | gteMAC2 = gteIR2 * gteIR2; | |
1028 | gteMAC3 = gteIR3 * gteIR3; | |
1029 | } | |
1030 | ||
1031 | void gteSQR_part_shift(psxCP2Regs *regs) { | |
1032 | gteFLAG = 0; | |
1033 | ||
1034 | gteMAC1 = (gteIR1 * gteIR1) >> 12; | |
1035 | gteMAC2 = (gteIR2 * gteIR2) >> 12; | |
1036 | gteMAC3 = (gteIR3 * gteIR3) >> 12; | |
1037 | } | |
1038 | ||
1039 | void gteOP_part_noshift(psxCP2Regs *regs) { | |
1040 | gteFLAG = 0; | |
1041 | ||
1042 | gteMAC1 = (gteR22 * gteIR3) - (gteR33 * gteIR2); | |
1043 | gteMAC2 = (gteR33 * gteIR1) - (gteR11 * gteIR3); | |
1044 | gteMAC3 = (gteR11 * gteIR2) - (gteR22 * gteIR1); | |
1045 | } | |
1046 | ||
1047 | void gteOP_part_shift(psxCP2Regs *regs) { | |
1048 | gteFLAG = 0; | |
1049 | ||
1050 | gteMAC1 = ((gteR22 * gteIR3) - (gteR33 * gteIR2)) >> 12; | |
1051 | gteMAC2 = ((gteR33 * gteIR1) - (gteR11 * gteIR3)) >> 12; | |
1052 | gteMAC3 = ((gteR11 * gteIR2) - (gteR22 * gteIR1)) >> 12; | |
1053 | } | |
1054 | ||
1055 | void gteDCPL_part(psxCP2Regs *regs) { | |
1056 | s32 RIR1 = ((s32)gteR * gteIR1) >> 8; | |
1057 | s32 GIR2 = ((s32)gteG * gteIR2) >> 8; | |
1058 | s32 BIR3 = ((s32)gteB * gteIR3) >> 8; | |
1059 | ||
1060 | gteFLAG = 0; | |
1061 | ||
1062 | gteMAC1 = RIR1 + ((gteIR0 * limB1(A1U((s64)gteRFC - RIR1), 0)) >> 12); | |
1063 | gteMAC2 = GIR2 + ((gteIR0 * limB1(A2U((s64)gteGFC - GIR2), 0)) >> 12); | |
1064 | gteMAC3 = BIR3 + ((gteIR0 * limB1(A3U((s64)gteBFC - BIR3), 0)) >> 12); | |
1065 | } | |
1066 | ||
1067 | void gteGPF_part_noshift(psxCP2Regs *regs) { | |
1068 | gteFLAG = 0; | |
1069 | ||
1070 | gteMAC1 = gteIR0 * gteIR1; | |
1071 | gteMAC2 = gteIR0 * gteIR2; | |
1072 | gteMAC3 = gteIR0 * gteIR3; | |
1073 | } | |
1074 | ||
1075 | void gteGPF_part_shift(psxCP2Regs *regs) { | |
1076 | gteFLAG = 0; | |
1077 | ||
1078 | gteMAC1 = (gteIR0 * gteIR1) >> 12; | |
1079 | gteMAC2 = (gteIR0 * gteIR2) >> 12; | |
1080 | gteMAC3 = (gteIR0 * gteIR3) >> 12; | |
1081 | } | |
1082 | ||
1083 | #endif // !FLAGLESS | |
1084 | ||
1085 | void gteGPL_part_noshift(psxCP2Regs *regs) { | |
1086 | gteFLAG = 0; | |
1087 | ||
1088 | gteMAC1 = A1((s64)gteMAC1 + (gteIR0 * gteIR1)); | |
1089 | gteMAC2 = A2((s64)gteMAC2 + (gteIR0 * gteIR2)); | |
1090 | gteMAC3 = A3((s64)gteMAC3 + (gteIR0 * gteIR3)); | |
1091 | } | |
1092 | ||
1093 | void gteGPL_part_shift(psxCP2Regs *regs) { | |
1094 | gteFLAG = 0; | |
1095 | ||
1096 | gteMAC1 = A1((s64)gteMAC1 + ((gteIR0 * gteIR1) >> 12)); | |
1097 | gteMAC2 = A2((s64)gteMAC2 + ((gteIR0 * gteIR2) >> 12)); | |
1098 | gteMAC3 = A3((s64)gteMAC3 + ((gteIR0 * gteIR3) >> 12)); | |
1099 | } | |
1100 | ||
1101 | void gteDPCS_part_noshift(psxCP2Regs *regs) { | |
1102 | int shift = 0; | |
1103 | ||
1104 | gteFLAG = 0; | |
1105 | ||
1106 | gteMAC1 = ((gteR << 16) + (gteIR0 * limB1(A1U((s64)gteRFC - (gteR << 4)) << (12 - shift), 0))) >> 12; | |
1107 | gteMAC2 = ((gteG << 16) + (gteIR0 * limB2(A2U((s64)gteGFC - (gteG << 4)) << (12 - shift), 0))) >> 12; | |
1108 | gteMAC3 = ((gteB << 16) + (gteIR0 * limB3(A3U((s64)gteBFC - (gteB << 4)) << (12 - shift), 0))) >> 12; | |
1109 | } | |
1110 | ||
1111 | void gteDPCS_part_shift(psxCP2Regs *regs) { | |
1112 | int shift = 12; | |
1113 | ||
1114 | gteFLAG = 0; | |
1115 | ||
1116 | gteMAC1 = ((gteR << 16) + (gteIR0 * limB1(A1U((s64)gteRFC - (gteR << 4)) << (12 - shift), 0))) >> 12; | |
1117 | gteMAC2 = ((gteG << 16) + (gteIR0 * limB2(A2U((s64)gteGFC - (gteG << 4)) << (12 - shift), 0))) >> 12; | |
1118 | gteMAC3 = ((gteB << 16) + (gteIR0 * limB3(A3U((s64)gteBFC - (gteB << 4)) << (12 - shift), 0))) >> 12; | |
1119 | } | |
1120 | ||
1121 | void gteINTPL_part_noshift(psxCP2Regs *regs) { | |
1122 | gteFLAG = 0; | |
1123 | ||
1124 | gteMAC1 = ((gteIR1 << 12) + (gteIR0 * limB1(A1U((s64)gteRFC - gteIR1), 0))); | |
1125 | gteMAC2 = ((gteIR2 << 12) + (gteIR0 * limB2(A2U((s64)gteGFC - gteIR2), 0))); | |
1126 | gteMAC3 = ((gteIR3 << 12) + (gteIR0 * limB3(A3U((s64)gteBFC - gteIR3), 0))); | |
1127 | } | |
1128 | ||
1129 | void gteINTPL_part_shift(psxCP2Regs *regs) { | |
1130 | gteFLAG = 0; | |
1131 | ||
1132 | gteMAC1 = ((gteIR1 << 12) + (gteIR0 * limB1(A1U((s64)gteRFC - gteIR1), 0))) >> 12; | |
1133 | gteMAC2 = ((gteIR2 << 12) + (gteIR0 * limB2(A2U((s64)gteGFC - gteIR2), 0))) >> 12; | |
1134 | gteMAC3 = ((gteIR3 << 12) + (gteIR0 * limB3(A3U((s64)gteBFC - gteIR3), 0))) >> 12; | |
1135 | } | |
1136 | ||
1137 | void gteMACtoRGB(psxCP2Regs *regs) { | |
1138 | gteRGB0 = gteRGB1; | |
1139 | gteRGB1 = gteRGB2; | |
1140 | gteCODE2 = gteCODE; | |
1141 | gteR2 = limC1(gteMAC1 >> 4); | |
1142 | gteG2 = limC2(gteMAC2 >> 4); | |
1143 | gteB2 = limC3(gteMAC3 >> 4); | |
1144 | } | |
1145 |